playing videos in JATOS from lab.js file

LiesbethGijbels

Hi there!

I wrote an audiovisual experiment in lab.js.

I want to do a pilot study by using the free server of JATOS (Thank you for that! ). When I upload my study and run it, the video's are delayed. It gives no video at all, or a black screen first and then shows the video, but my audio and video are separate (and it shows synchronized in lab.js without the black screen) so audio and video are completely off when I run it in JATOS. Does anyone have experience with this or knows how to solve it?

It runs without problems in lab.js.

The videos are .mov files, but I also tried it with a .mp4 file, that doesn't change anything.

There are 20 different videos, but they are all relatively small (between 500-900kb). The videos are integrated in the .json file

I also tried to make a very basic experiment to see if I could just get a video to run. So i made a lab.js file with a start screen and then 1 video, but this has the same problems. (I wanted to upload the zipfile here, downloaded from lab.js but it gives me error 431?)

I tried to run it in chrome and Safari, both have the same problems.

It would be amazing if someone could help me with this issue! Thanks!

Liesbeth Gijbels

kri

Hi Liesbeth,

experiements.jatos.org should be able to handle your experiment as far as I can see: 20 different videos each 500-900kb is not too big.

the video's are delayed. It gives no video at all, or a black screen first and then shows the video, but my audio and video are separate (and it shows synchronized in lab.js without the black screen) so audio and video are completely off when I run it in JATOS

It sounds like network delay to me. As soon as you run the study in the internet those delays or other network problems can happen. What happens when you try to run your study in a local JATOS installation? Locally you won't have those internet problems.

Btw. I guess you get the 431 error code because it's a zip file. The same happen with Gmail. There could be dangerous stuff inside a zip. One way around this is to use Google Drive or similar services.

Best,

Kristian

LiesbethGijbels

Hi Kristian,

Thank you for your response.

It works perfect when I run it locally.

So how do I solve this? What is the best way to get my experiment out there? I made the experiment online because of COVID, so running it on my computer gives me the same problem as I had before, when I wrote it in Python. I can't share it with anyone?

Thank you for your response

Liesbeth

kri

Hi Liesbeth,

If you post/send the study I can have a look at it. But actually it would be better to ask someone from lab.js, since this is not so much a JATOS problem.

Best,

Kristian

LiesbethGijbels

Hi Kristian,

Thank you.

I am also talking to lab.js, but they had sent me to you.

It would be great though if you could have a look at it: https://drive.google.com/drive/folders/1TKW3iLxrS_HkJnv280E5759fy2dFtRpX?usp=sharing

Thank you so much

Liesbeth

kri

Hi Liesbeth,

I see in your study that you load the audio and video files just before they are played. It would be better to pre-load them at the beginning of the study. Loading larger files like audio or video always take some time and here in your experiment likely take different times but are played right away which leads to not synchronized audio and video and lagging. If you pre-load them, they are already in your participant's browser memory and can be played right away. Unfortunately my knowledge of lab.js is not good enough to help your with pre-loading.

Another thing that I noticed but this might be purpose: your audio files have quite some background noise. Is this maybe part of your experiment?

And yet another thing: it would be easier if you combine your video and audio files into one file. This would make audio and video synchronization way easier. But somehow I suspect you are aware of this and just need to play them separately in different combinations in your experiment.

Best,

Kristian

LiesbethGijbels

Hi Kristian,

Thank you for this!

I will try to preload!

It is a speech in noise task, so yes indeed the background noise is part of the experiment. And I would rather not combine them for the exact reason that I want different combinations and now it needs to load a small amount of different videos but a large amount of audio, which is "less heavy", If I combine them there will be a lot of 'heavier' video files, so I was trying to avoid that.

Liesbeth Gijbels